EP0788290A1 - Programmable hearing aid - Google Patents

Abstract

The programmable hearing aid (1) comprises an amplification and transmission unit (4) in a path from a microphone (2) to a speaker (3) which has adjustable transmission characteristics. Signals (6) are sampled from one or more points (5) of the signal path and supplied to a module (7) for transformation of the signals from the time domain into a frequency domain. The frequency domain transformed signals are supplied to a second module (9) for evaluation. The second module produces control signals (10), supplied to the amplification and transmission unit, which are used to select one or more parameters of the transmission path, or to change the transmission characteristics. The parameters are stored in a data memory (11) associated with the signal path.

Description

The invention relates to a programmable hearing aid with an amplifier and transmission part which can be adjusted in terms of its transmission properties between the microphone and the listener to different transmission characteristics.

In a hearing device known from EP-B-0 064 042, eight parameter sets for different transmission characteristics for different environmental situations are stored in a hearing device memory. The various parameter sets for the eight stored programs can be called up one after the other by pressing a switch. A control unit controls a signal processor connected between the microphone and the handset, which then sets a first transmission function intended for an intended environmental situation. However, the stored signal transmission programs can only be called up one after the other via the switch until, in the opinion of the hearing device wearer, the transmission function that is just right for the given environmental situation has been found. Accordingly, a plurality of parameter sets, so-called hearing situations, which can be selected by the user, are generally stored in the known programmable hearing aid. Each of these parameter sets represents the sensibly coordinated setting of all signal processing parameters for a specific acoustic situation, e.g. at rest, i.e. without disturbing background noise or a conversation situation with low-frequency noise, etc. The hearing aid wearer selects the desired situation by pressing a button on the hearing aid.

EP-A-0 674 464 relates to a programmable hearing aid with a fuzzy logic controller. This hearing aid has a control system, it being used for automatic switching and adjustment to the respective environmental situation has a controller assigned to the amplifier and transmission part which, depending on the input variables characterizing the respective environmental situation, makes a selection of the parameter sets stored in the data carrier of the hearing aid or of parameters for changing transmission characteristics of the hearing aid.

For an automatic or largely automatic adaptation of a programmable hearing aid dependent on input or measurement signals, a neuronal structure which is subordinate to the signal path from the microphone to the listener and is assigned to a data carrier is proposed in European patent application 94 117 797.4-2211, with a data carrier being assigned, with the signal path being used signals are tapped from one or more tapping points and fed to a module for signal processing, and the processed signals can be fed to the neural structure, which generates control signals for selecting parameters of the amplifier and transmission part stored in a data memory assigned to the signal path or for changing the amplifier - And transmission characteristics can be delivered to the amplifier and transmission part.

The proposals according to EP-A-0 674 464 and EP 94 117 797.4-2211 relate to the determination of hearing programs and signal processing parameters by means of fuzzy logic or by means of a neural structure, whereby in both cases an evaluation of the input signal (s) (see ) in the time domain and the realization of the fuzzy logic or the neural structure is given in analog circuitry.

The object of the invention is to provide a hearing device of the type mentioned at the outset, which enables a largely automatic selection of a respective hearing program or the automatic setting of individual signal processing parameters by evaluation of the acoustic signal arriving in the hearing aid, whereby special possibilities of signal evaluation should be usable.

According to the invention, this object is achieved in that signals are tapped from the signal path from one or more tapping points and fed to a module for transforming the signals from the time domain into the frequency domain, the signals transferred into the frequency domain being fed to a further module for signal evaluation in the frequency domain, wherein this module generates control signals which can be output to the amplifier and transmission part in order to select parameters of the amplifier and transmission part stored in a data memory assigned to the signal path or to change the amplifier and transmission characteristic.

In the hearing aid according to the invention, a signal evaluation is provided, according to which the incoming acoustic signal is first subjected to a spectral analysis and the actual analysis then takes place in the frequency domain. The algorithms and rule sets for the analysis of the signal that has been preprocessed in this way should be as flexible as possible or programmable. The digital circuit implementation offers special advantages. For the logical implementation of the algorithms and rule sets there are alternative mathematical calculation rules, fuzzy logic and neural structures. According to the invention, a better adaptation of the signal processing to the hearing damage is possible through continuous signal-dependent setting of the signal processing parameters. At the same time, the hearing device wearer is relieved of having to switch hearing programs themselves.

Advantageous embodiments of the invention are characterized by the claims.

Further advantages and details of the invention are explained in more detail below with reference to the exemplary embodiments shown in the figures.

• Figur 1 ein Blockschaltbild eines erfindungsgemäßen Hörgerätes mit einem Verstärker- und Übertragungsteil für eine analoge Signalverarbeitung,
• Figur 2 ein Blockschaltbild eines erfindungsgemäßen Hörgerätes mit einem Verstärker- und Übertragungsteil für eine digitale Signalverarbeitung,
• Figur 3 ein Blockschaltbild eines erfindungsgemäßen Hörgerätes mit einem Verstärker- und Übertragungsteil für eine digitale Signalverarbeitung, wobei das Hörgerät-Modul zur Signalauswertung im Frequenzbereich mit einem Speicher im Datenaustausch steht,
• Figuren 4 bis 6 Blockschaltbilder für mögliche Hörgerät-Module zur Signalauswertung im Frequenzbereich und zur Bestimmung des - in Abhängigkeit vom Hörgerät empfangenen akustischen Signals - jeweils zu aktivierenden Hörprogramms,
• Figuren 7 bis 9 Blockschaltbilder für mögliche Hörgerät-Module zur Signalauswertung im Frequenzbereich und zur Bestimmung einzelner Signalverarbeitungsparameter, in Abhängigkeit des vom Hörgerät empfangenen akustischen Signals.

Show it:

• FIG. 1 shows a block diagram of a hearing aid according to the invention with an amplifier and transmission part for analog signal processing,
• FIG. 2 shows a block diagram of a hearing device according to the invention with an amplifier and transmission part for digital signal processing,
• FIG. 3 shows a block diagram of a hearing aid according to the invention with an amplifier and transmission part for digital signal processing, the hearing aid module being in data exchange with a memory for signal evaluation in the frequency range,
• FIGS. 4 to 6 block diagrams for possible hearing aid modules for signal evaluation in the frequency range and for determining the acoustic program to be activated, depending on the hearing aid received,
• Figures 7 to 9 block diagrams for possible hearing aid modules for signal evaluation in the frequency domain and for determining individual signal processing parameters, depending on the acoustic signal received by the hearing aid.

The hearing aid 1 according to the invention, shown schematically in FIG. 1, records 2 sound signals via a microphone. This acoustic information is converted into electrical signals in the microphone. After signal processing in an amplification and transmission part 4, the electrical signal is fed to a receiver 3 as an output converter. With the hearing aid According to FIG. 1, analog signal processing is provided in the amplifier and transmission part 4. 5 analog signals 6 are tapped from the signal path from one or more tapping points.

A module 12 for selecting the tap signal 13 determines from which point in the signal path the signal is tapped and then feeds it to an analog / digital converter 14. It can make this selection either according to its own algorithm implemented in it or according to an external control signal 18. The signal 15 digitized in the analog / digital converter 14 is then fed to a module 7 for transforming the signal (s) 15 from the time domain into the frequency domain. The data or signals 8 transformed in module 7 are then further processed by a module 9 for signal evaluation in the frequency domain. The essential signal analysis and signal recognition algorithms run here, system information from the hearing aid also being able to be included in the selection decision when determining the new signal processing parameters; e.g. via the position of operating elements or the respective battery charge state and by means 16 for detecting system states and by means of its output signals 17. The control signals 10 generated in module 9 are used to select parameters of the amplifier and transmission part or stored in a data memory assigned to the signal path deliverable to the amplifier and transmission part 4 for changing the amplifier and transmission characteristics. According to FIG. 1, control signals 18 of a control module 19 can be fed to the module 12 for the selection of one or more tap signals 13, whereby the selection of the tap signals can be influenced.

If, according to the exemplary embodiment in FIG. 2, digital signal processing takes place in the signal path in a digital hearing device 1 'in the amplifier and transmission part 4', no separate analog / digital converter is required for signal analysis. Instead, the already digital signal 6 'can be tapped from the signal path and processed further for the signal analysis. The tapped digital signal 6 'is fed after the module 12 to select one or more tapping signals as a digital tapping signal 13' to the module 7 for transforming the signal into the frequency range. The further signal analysis can proceed as described above.

The transformation of the tapped signal into the frequency range occurs e.g. according to the known methods of discrete Fourier transform or Fast Fourier transform. As a result of the transformation, the amplitude and phase values of the signal are available at a selectable number of support points.

Together with the circuit diagrams in FIGS. 3 to 6, the determination of the hearing program by means of signal evaluation in the frequency range is described below, specifically the determination of the hearing program by means of a mathematical calculation rule, via a fuzzy logic rule set or with the aid of a neural structure. The acoustic environment must first be analyzed and classified. It will generally result that it corresponds to a mixture of the listening situations on which the hearing programs are based (= corner hearing situations). Accordingly, it must be decided which of the stored hearing programs should be activated or how a suitable mean hearing program should be obtained from the stored hearing programs.

According to an advantageous embodiment according to FIG. 3, the module 9 for signal evaluation in the frequency domain is in data exchange with a memory 20, the module 9 emitting signals 21 to the memory 20 and causing the memory to send certain data 22 to the amplifier and transmission part 4 to deliver or take data 23 from the memory 20, process and deliver them as control signals 10 to the amplifier and transmission part 4.

To determine the listening situation via mathematical functions, the module 9 for signal evaluation in the frequency domain, as shown in FIG. 4, has a component 24 for normalizing the amplitude values, a component 25 for forming frequency-related parameters, a component 26 for forming time parameters, and a component 27 to form similarity measures and a component 28 for determining the respective hearing program to be activated.

Standardization:
Since the situation analysis should be independent of the absolute loudness of the signal to be examined, the amplitude values are normalized in the first processing step. One way to do this is to determine the factor by which the greatest amplitude value must be multiplied in order to assume a predetermined maximum value. Then all amplitude values are multiplied by this factor.

Formation of frequency-related parameters:
Since it is not the amplitude values alone that are required for the further evaluation, but rather parameters that describe their relationship to one another, these must be formed. Examples for this are:
Sum of amplitude values.
Differences in amplitude values.
Mean values of amplitude values. These can also be weighted.
Power of the signal in certain frequency ranges.
Average rise / fall of the amplitude values in certain frequency ranges.

The exact calculation of these intermediate quantities required is part of the overall analysis algorithm.

Creation of time parameters:
Since, in addition to the variables available up to now, each of which is valid for a time interval of the Fourier transformation, their temporal course is also important for the signal analysis, additional temporal parameters are then determined from the previous parameters and amplitude values.
Examples for this are:
Totals, sometimes also weighted, of parameters at different times.
Differences, sometimes also weighted, of parameters at different times.
Average values of parameters at different times.
Average increase / decrease in parameters within a certain time interval.

• Kennwertsatz von Eck-Hörsituation 1: W1 = ( w11, w12, w13)
• Kennwertsatz von Eck-Hörsituation 2: W2 = ( w21, w22, w23)
• Kennwertsatz von Eck-Hörsituation 3: W3 = ( w31, w32, w33)
• Kennwertsatz der aktuellen Hörsituation: WA = (wA1, wA2, wA3)

Berechnung der Ähnlichkeitsmaße:

• Ähnlichkeitsmaß 1: S1 = g1 * | w11 - wA1 | + g2 * | w12 -wA2| + g3 * | w13 - wA3 |
• Ähnlichkeitsmaß 2: S2 = g1 * | w21 - wA1 | + g2 * | w22 -wA2| + g3 * | w23 - wA3 |
• Ähnlichkeitsmaß 3: S3 = g1 * | w31 - wA1 | + g2 * | w32 -wA2| + g3 * | w33 - wA3 |
• Hierbei bedeutet | ... | die mathematische Funktion der Betragsbildung und
• g1, g2, g3 stellen die Gewichtungsfaktoren der einzelnen Kennwerte dar.

Calculation of similarity measures:
The next step is to determine how similar the current acoustic situation is to the given listening situations (= corner hearing situations).
For this purpose, the deviation of its characteristic value set from the currently determined characteristic value set of the acoustic ambient situation is determined for each corner hearing situation. For example, for each situation comparison, the absolute value of the difference is determined for each characteristic value and weighted with a factor to be specified. The sum of these weighted differences gives a measure of the similarity of the current situation with the respective corner hearing situation. The weighting of the differences leads to the fact that deviations of different characteristic values are more or less included in the calculation of the similarity measure depending on their meaning.
This procedure can easily be described mathematically, as shown in the following example. 3 corner hearing situations with 3 characteristic values each are assumed.

• Characteristic set of corner hearing situation 1: W1 = (w11, w12, w13)
• Characteristic set of corner hearing situation 2: W2 = (w21, w22, w23)
• Characteristic set of corner hearing situation 3: W3 = (w31, w32, w33)
• Set of characteristic values of the current hearing situation: WA = (wA1, wA2, wA3)

Calculation of similarity measures:

• Similarity measure 1: S1 = g1 * | w11 - wA1 | + g2 * | w12 -wA2 | + g3 * | w13 - wA3 |
• Similarity measure 2: S2 = g1 * | w21 - wA1 | + g2 * | w22 -wA2 | + g3 * | w23 - wA3 |
• Similarity measure 3: S3 = g1 * | w31 - wA1 | + g2 * | w32 -wA2 | + g3 * | w33 - wA3 |
• Here, | means ... | the mathematical function of the amount formation and
• g1, g2, g3 represent the weighting factors of the individual parameters.

• Hörprogramm 1: P1 = ( p11, p12, p13 )
• Hörprogramm 2: P2 = ( p21, p22, p23 )
• Hörprogramm 3: P3 = ( p31, p32, p33 )

Zu berechnendes Hörprogramm: PN = ( pN1, pN2, pN3 ) Berechnung der Signalverarbeitungsparameter des einzustellenden Hörprogramms: $$\begin{array}{c}\begin{array}{c}\text{pN1 =}\frac{\text{(p11 * S1) + (P21 * S2) + (P31 * S3)}}{\text{S1 + S2 + S3}}\\ \text{pN2 =}\frac{\text{(p12 * S1) + (P22 * S2) + (P32 * S3)}}{\text{S1 + S2 + S3}}\\ \text{pN3 =}\frac{\text{(p13 * S1) + (P23 * S2) + (P33 * S3)}}{\text{S1 + S2 + S3}}\end{array}\end{array}$$

Determination of the hearing program to be activated:
Now that there is a similarity measure for the current acoustic situation with every corner hearing situation, the hearing program to be activated must be determined. In the simplest case, the hearing program is activated for which the degree of similarity between the corresponding corner situation and the current acoustic situation has the greatest value. Here, however, a switching hysteresis must be provided in order to avoid constant switching between two hearing programs in borderline situations.
In addition to this simple solution, there is the possibility of calculating the hearing program to be activated from the stored hearing programs in such a way that an average is formed for each parameter, which reflects the individual measures of similarity between the current situation and corner situations.
For example, this could be done by multiplying each parameter of the corner situations by the corresponding measure of similarity, then adding these quantities and then the value obtained is divided by the sum of the similarity measures. This would result in a kind of "interpolated average" for each parameter.
This method can also be easily described mathematically, as shown in the following example. It is based on 3 hearing programs with 3 signal processing parameters each.

• Hearing program 1: P1 = (p11, p12, p13)
• Hearing program 2: P2 = (p21, p22, p23)
• Hearing program 3: P3 = (p31, p32, p33)

Hearing program to be calculated: PN = (pN1, pN2, pN3) Calculation of the signal processing parameters of the hearing program to be set: $$\begin{array}{c}\begin{array}{c}\text{pN1 =}\frac{\text{(p11 * S1) + (P21 * S2) + (P31 * S3)}}{\text{S1 + S2 + S3}}\\ \text{pN2 =}\frac{\text{(p12 * S1) + (P22 * S2) + (P32 * S3)}}{\text{S1 + S2 + S3}}\\ \text{pN3 =}\frac{\text{(p13 * S1) + (P23 * S2) + (P33 * S3)}}{\text{S1 + S2 + S3}}\end{array}\end{array}$$

It can be advantageous to implement the calculation of the similarity measures using fuzzy logic instead of a mathematical implementation. The advantages of fuzzy logic compared to a closed mathematical approach are that even non-linear relationships can be easily described and edited, such as those that can occur on the edge of permissible value ranges. Even mutually contradicting values of input variables can be meaningfully processed together. As described above, the steps of standardization, formation of frequency-related parameters and formation of temporal ones are also used for this purpose Characteristics accomplished. To calculate the similarity measures, the relevant amplitude values and parameters are now subjected to the steps of fuzzification, inference and defuzzification. As a result, similarity measures are also available that describe the similarity to the current acoustic environment for each corner hearing situation. The hearing program to be activated is then determined in the same way as described above.

To determine the hearing situation via fuzzy logic, a module 9 for signal evaluation in the frequency range is provided in accordance with FIG Component 29 for fuzzification, a component 30 for inference formation, a component 31 for defuzzification of the similarity measures and a component 28 for determining the respective hearing program to be activated.

To determine the hearing situation via a neural structure, a module 9 for signal evaluation in the frequency range is provided according to the invention, consisting of a component 24 for normalizing the amplitude values, a component 25 for forming frequency-related parameters, a component 26 for forming time parameters and a component 32 for realizing a neural structure for determining the respective hearing program to be activated. 6, the module 9 can emit signals 21 to a memory 20, which then supplies data signals 22 to the amplifier and transmission part 4 for determining the hearing program to be activated in each case.

gelerntes" Entscheidungswissen zu implementieren, selbst wenn dieses nicht explizit formuliert werden kann. Hierzu werden, wie bereits beschrieben, ebenfalls die Schritte Normierung, Bildung von frequenzbezogenen Kenngrößen und Bildung von zeitlichen Kenngrößen vollzogen. Die Amplitudenwerte und Zwischenvariablen werden nun auf die Eingänge einer neuronalen Struktur gegeben. Entsprechend dem in seiner Struktur festgelegten Übertragungsverhalten stellen die an seinen Ausgängen dann anliegenden Werte einen Code dar, der das zu aktivierende Hörprogramm auswählt. Eine weitere Berechnung des zu aktivierenden Hörprogramms, wie bei den Ausführungen der Figuren 4 und 5, ist in diesem Fall also nicht erforderlich.It can be advantageous to use a neural structure to determine the hearing program because it is then possible is also empirically gained

to implement learned decision-making knowledge, even if this cannot be formulated explicitly. For this purpose, as already described, the steps of standardization, formation of frequency-related parameters and formation of time parameters are also carried out. The amplitude values and intermediate variables are now applied to the inputs of a neural structure In accordance with the transmission behavior defined in its structure, the values present at its outputs then represent a code which selects the hearing program to be activated, in which case a further calculation of the hearing program to be activated is carried out, as in the embodiments of FIGS so not necessary.

Instead of determining a complete set of signal processing parameters (hearing program), each parameter can also be determined individually. The determination of individual signal processing parameters by signal evaluation in the frequency domain is described below with reference to the circuit diagrams of FIGS. 7 to 9, namely by determining the parameters via mathematical functions, via fuzzy logic or via a neural structure.

In order to determine the signal processing parameters according to a mathematical calculation rule, the amplitude values must also be normalized, as already described. The value is then calculated in a component 33 for determining the individual signal processing parameters for each parameter to be determined according to a formula, for example according to:

G₂

den zu berechnenden Parameter, hier z.B. die Verstärkung im Kanal 2.

A_i(...)

den i-ten Amplitudenwert der Fourieranalyse.

(n), (n-1), ...

kennzeichnet, daß es sich um den jeweiligen Wert im Zeitintervall n, n-1, ... der Fourieranalyse handelt.

Here means:

G ₂

the parameter to be calculated, here for example the gain in channel 2.

A _i (...)

the i-th amplitude value of the Fourier analysis.

(n), (n-1), ...

indicates that it is the respective value in the time interval n, n-1, ... of the Fourier analysis.

It may be advantageous to use fuzzy logic to calculate the signal processing parameters instead of using a mathematical rule. The advantages of fuzzy logic compared to a closed mathematical approach are that even non-linear relationships can be easily described and edited, such as those that can occur on the edge of permissible value ranges. Even mutually contradicting values of input variables can be meaningfully processed together. For this purpose, the steps of standardization 24, formation of frequency-related parameters 25 and formation of time parameters 26 are also carried out. To calculate the parameter values, however, the relevant amplitude values and parameters are subjected to the steps of fuzzification, inference and defuzzification using the module components 34, 35, 36. The required parameter values are then available as a result.

Finally, it can be advantageous to use a neural structure to determine the parameters. This makes it possible to implement empirically acquired, learned decision knowledge, even if this cannot be formulated explicitly. As described, the steps of standardization, formation of frequency-oriented parameters and formation of time parameters are also carried out for this purpose. The amplitude values and intermediate variables are then given in a module component 37 to the inputs of a neural structure. In accordance with the transmission behavior defined in its structure, the values then applied to its outputs represent the setting values of the signal processing parameters.

Claims (14)

Programmable hearing aid (1, 1 ') with an amplifier and transmission part (4, 4') adjustable in its transmission properties between the microphone (2) and receiver (3) to different transmission characteristics, characterized in that the signal path from one or more tapping points (5) Signals (6, 6 ') are tapped and fed to a module (7) for transforming the signals from the time domain into the frequency domain, the signals (8) transferred into the frequency domain being fed to a further module (9) for signal evaluation in the frequency domain This module (9) generates control signals (10) which are used to select parameters of the amplifier and transmission part stored in a data memory (11) assigned to the signal path or to change the amplifier and transmission characteristics to the amplifier and transmission part (4 ) are dispensable. Hearing aid according to claim 1, characterized in that a module (12) for selecting one or more tap signals is provided between the tap point (5) of the signals (6, 6 ') and the module (7) for transforming the signals from the time domain into the frequency domain is. Hearing aid according to claims 1 and 2, characterized in that in an amplifier and transmission part (4) for analog signal processing, analog signals (6) are tapped from one or more tapping points (5), the module (12) for selecting one or more tapping signals fed and then the specific tap signals (13) via an analog / digital converter (14) and then as digital signals (15) can be fed to the module (7) for transforming the signals from the time domain to the frequency domain. Hearing aid according to claims 1 and 2, characterized in that in an amplifier and transmission part (4 ') for digital signal processing, digital signals (6') are tapped from one or more tapping points (5), the module (12) for selecting one or several tap signals and then the specific digital tap signals (13 ') can be fed directly to the module (7) for transforming the signals from the time domain to the frequency domain. Hearing aid according to one or more of claims 1 to 4, characterized in that means (16) for detecting system states of the hearing aid (1, 1 ') are provided, the output signals (17) of which can be fed to the module (9) for signal evaluation in the frequency range and These output signals (17) can be taken into account when generating the control signals (10). Hearing aid according to one or more of Claims 1 to 5, characterized in that control signals (18) from a control module (19) can be fed to the module (12) for selecting one or more tap signals (13, 13 ') and the selection of the tap signals can thereby be influenced is. Hearing aid according to claim 1, characterized in that the transformation of the + signals from the time domain to the frequency domain can be carried out in a manner known per se after the discrete Fourier transform or the Fast Fourier transform or after a discrete cosine transform or a discrete wavelet transform. Hearing aid according to one or more of claims 1 to 7, characterized in that the Module (9) for signal evaluation in the frequency domain is in data exchange with a memory (20), the module (9) emitting signals (21) to the memory (20) and causing the memory to send certain data (22) to the amplifier. and to transmit the transmission part (4) or take data (23) from the memory (20), process them and deliver them as control signals (10) to the amplifier and transmission part (4) (FIG. 3). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values, a component (25) for forming frequency-related parameters, a component (26) to form time parameters, a component (27) for calculating similarity measures and a component (28) for determining the respective hearing program to be activated (FIG. 4). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values, a component (25) for forming frequency-related parameters, a component (26) for the formation of time parameters, a component (29) for fuzzification, a component (30) for inference formation, a component (31) for defuzzification of the similarity measures and a component (28) for determining the respective hearing program to be activated (FIG. 5). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values, a component (25) for forming frequency-related parameters, a component (26) for the formation of time parameters and has a component (32) for realizing a neural structure for determining the respective hearing program to be activated (FIG. 6). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values and a component (33) for determining the individual signal processing parameters according to a mathematical calculation specification ( Figure 7). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values, a component (25) for forming frequency-related parameters, a component (26) to form time parameters, a component (34) for fuzzification, a component (35) for inference formation and a component (36) for defuzzification of the individual signal processing parameters to be output (FIG. 8). Hearing aid according to one or more of claims 1 to 8, characterized in that the module (9) for signal evaluation in the frequency range has a component (24) for normalizing the amplitude values, a component (25) for forming frequency-related parameters, a component (26) for the formation of time parameters and a component (37) for realizing a neural structure for determining the individual signal processing parameters (FIG. 9)

Images